CHESS is an interactive health communication system that has successfully improved quality of life, information skills, and health care utilization for cancer patients. Further advancing interactive health communication systems will require integrating them with human-based systems that already do or potentially could serve patients as well. This project will investigate the benefits continuing access to a human Cancer Mentor who keeps track of the patient and calls her regularly to offer help, thus potentially providing patients with considerable tailoring and guidance. Those benefits should be even more effective in the company of an Interactive Cancer Communication System like CHESS, because the mentor will know what CHESS will know about the patient as well. Conversely, the Cancer Mentor's knowledge of CHESS and her recommendations for how to use it should lead patients to use it more effectively as well. We will therefore employ a fully-crossed design to test the efficacy of CHESS and a human Cancer Mentor, against a control condition in which patients are given only access to the Internet (rapidly becoming a de facto standard information care situation). Additionally, we will test the benefits of combining the two against those of either alone. With a sample of 360 breast cancer patients from three hospitals randomized to the four arms, we will measure the effect of these conditions on the primary outcomes of five aspects of patient quality of life, and cost of providing the interventions. Secondary analyses will examine mediating processes such as satisfaction with the medical system, health self-efficacy (including information competence), knowledge of their own condition, sense of an expert watching over them (and trust in those experts), and information overload.